Hardware-Aware Color-Distorted Image Classifier for Intelligent White Balance
International Journal of Image and Graphics(2025)
Zunyi Normal Univ
Abstract
Automatic white balance (AWB) is an important module for cameras and classification of the color-distorted image is critical to realize intelligent AWB. Though accurate classifiers usually can be achieved via deep neural network models, they cannot fit into embedded hardware due to their complexity. To increase the classification accuracy and decrease latency, a lightweight convolutional neural network (CNN) with a histogram layer for AWB (AWBHNet) is constructed, which consists of one histogram layer, one regular convolutional layer, three depth separable convolutional layers, four pooling layers, two fully connected layers and two dropout layers. One-tenth of ImageNet is utilized as the normal image dataset. To generate various distorted colors, histogram shifting and matching are proposed to randomly adjust the histogram position or shape. Furthermore, the extent of shifting or matching is randomly generated to ensure the diversity of color distortion. Subsequently, the proposed AWBHNet and other CNNs are successively trained. Experiments show that the accuracy of the classifier trained by AWBHNet is 0.9150, which is at least 1.33% higher than each regular or lightweight network. Finally, intelligent AWB is realized on smartphones, the inference latency of AWBHNet is 47.6% lower than the best existing network.
MoreTranslated text
Key words
Automatic white balance,color-distorted image dataset,histogram layer,accuracy,latency
求助PDF
上传PDF
View via Publisher
AI Read Science
AI Summary
AI Summary is the key point extracted automatically understanding the full text of the paper, including the background, methods, results, conclusions, icons and other key content, so that you can get the outline of the paper at a glance.
Example
Background
Key content
Introduction
Methods
Results
Related work
Fund
Key content
- Pretraining has recently greatly promoted the development of natural language processing (NLP)
- We show that M6 outperforms the baselines in multimodal downstream tasks, and the large M6 with 10 parameters can reach a better performance
- We propose a method called M6 that is able to process information of multiple modalities and perform both single-modal and cross-modal understanding and generation
- The model is scaled to large model with 10 billion parameters with sophisticated deployment, and the 10 -parameter M6-large is the largest pretrained model in Chinese
- Experimental results show that our proposed M6 outperforms the baseline in a number of downstream tasks concerning both single modality and multiple modalities We will continue the pretraining of extremely large models by increasing data to explore the limit of its performance
Upload PDF to Generate Summary
Must-Reading Tree
Example

Generate MRT to find the research sequence of this paper
Related Papers
2008
被引用226 | 浏览
2018
被引用18 | 浏览
2017
被引用32568 | 浏览
2017
被引用11306 | 浏览
2016
被引用75 | 浏览
2020
被引用115 | 浏览
2021
被引用128 | 浏览
2020
被引用328 | 浏览
2023
被引用4 | 浏览
2022
被引用32 | 浏览
2024
被引用8 | 浏览
Data Disclaimer
The page data are from open Internet sources, cooperative publishers and automatic analysis results through AI technology. We do not make any commitments and guarantees for the validity, accuracy, correctness, reliability, completeness and timeliness of the page data. If you have any questions, please contact us by email: report@aminer.cn
Chat Paper